Abstract
A primary requirement in distributed robotic software systems is the dissemination of data to all interested collaborative entities in a timely and scalable manner. However, providing such a service in a highly dynamic and resource-limited robotic environment is a challenging task, and existing robot software infrastructure has limitations in this aspect. This paper presents a novel robot software infrastructure, micROS-drt, which supports real-time and scalable data distribution. The solution is based on a loosely coupled data publish-subscribe model with the ability to support various time-related constraints. And to realize this model, a mature data distribution standard, the data distribution service for real-time systems (DDS), is adopted as the foundation of the transport layer of this software infrastructure. By elaborately adapting and encapsulating the capability of the underlying DDS middleware, micROS-drt can meet the requirement of real-time and scalable data distribution in distributed robotic systems. Evaluation results in terms of scalability, latency jitter and transport priority as well as the experiment on real robots validate the effectiveness of this work.
Highlights
Consider the following robot-assisted urban search and rescue (USAR) [1] scenario
This scenario involves a distributed robotic system in which various kinds of data have to be shared among participants, such as the video captured by the robots and the control commands issued by the operators
This paper presents the design and implementation of micROS-drt, a software infrastructure that supports realtime and scalable data distribution in distributed robotic systems
Summary
Consider the following robot-assisted urban search and rescue (USAR) [1] scenario. A large-scale region should be explored after an earthquake to analyze the disaster situation and localize the victims. The human operators monitor the behavior of the robots and the video captured by them, analyze the collected data and guide the actions of the robots remotely. This scenario involves a distributed robotic system in which various kinds of data have to be shared among participants, such as the video captured by the robots and the control commands issued by the operators. This section discusses the requirements of distributed real-time computing in robotic settings firstly and introduces two software entities highly related to our work (i.e., DDS and ROS). Time-related constraints such as deadline or priority play significant roles in the logical correctness of a robot software system. The motivated scenario in “Introduction” section is a typical example of this concept
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
